This competing continuation proposal seeks support for a year of data analysis and report writing f or a randomized trial of two forms of group relapse prevention for cocaine addiction. The study recruited and randomly assigned 214 cocaine addicts to three experimental treatment conditions (a behavioral group model, a psychodynamic group model and a "no-group" control condition) . All subjects received individual drug abuse counseling. subjects committed to six months of treatment, and upon the six month anniversary were ref erred to another agency for less intensive aftercare. All but a handful became engaged in the experimental treatment, and sixty percent completed the entire six months. Group and individual treatment sessions were tape recorded, and clinicians completed reports of client outcomes for all treatment contacts. -Research staff collected comprehensive background and outcome data by interview from all randomized subjects just prior to randomization, at six months and twelve months after randomization. Subjects also gave urine samples at follow-up interviews. Six month follow-up data have been obtained on 94% of the 214, and twelve month follow-up data have been collected on 91% the 147 who have reached their 12 month anniversaries thus far. The last twelve month anniversaries are scheduled at the fifth month of year 05 of the current grant. The focus of this grant will be analysis of the 12 month follow-up data. The analysis will include comprehensive process and outcomes of treatment, and will focus on the long-term efficacy of the group conditions and any interaction between treatment models and client characteristics. The results will be presented in journal articles and a monograph.